Global Tech Governance: DeepSeek, OpenAI, and Anthropic Address UN Security Council on AI Safety
In a groundbreaking moment for global technology governance, the United Nations Security Council is preparing to host a high-level briefing on AI safety. This week, senior representatives from OpenAI, Anthropic, and — for the first time — Chinese AI startup DeepSeek are all expected to address the 15-member body. Scheduled during the UN General Assembly in New York, this meeting places AI risk at the forefront of the world’s most influential international security forum.
The briefing follows a series of unsettling AI incidents that have shaken policymakers worldwide. The UN-backed Independent International Scientific Panel on AI published its inaugural thematic brief on September 21. The panel warns that existing AI safety measures are “unravelling” and that recent agent breakouts raise fears humans may one day be unable to control or halt advanced AI systems. UN Secretary-General António Guterres issued a statement supporting the panel’s findings, urging frontier AI labs and safety institutes to engage further.
Anthropic CEO Dario Amodei has suggested embedding independent evaluators within frontier labs and coordinating a potential “speed limit” on AI self-improvement. Meanwhile, Google DeepMind CEO Demis Hassabis has advocated for an industry standards body modeled on the U.S. financial regulator FINRA. The participation of DeepSeek is viewed as particularly significant, marking the first time a Chinese frontier AI lab will formally address the Security Council on AI risk. This could potentially redefine the geopolitics of artificial intelligence governance.
Source: Inside AI – DeepSeek to Brief UN Security Council on AI Risks This Week